JOURNAL O F T H E AMERICAN CHEMICAL SOCIETY Registered i n U.S.Patent O&e.
@ Copyrighl. 1965, by the American Chemical Society
AUGUST5, 1965
VOLUME87, NUMBER 15
Physical and Inorganic Chemistry Reactions of Gaseous Ions. XIV. Mass Spectrometric Studies of Methane at Pressures to 2 Torr F. H. Field and M. S . B. Munson
Contribution from the Esso Research and Engineering Company, Baytown Research and Development Division, Baytown, Texas. Received February 15, 1965
Gaseous ionic reactions have been observed in the source of a mass spectrometer at pressures as high as 2 torr. The general pattern of reactions is the same as that observed previously at pressures up to 0.3 torr. CHj+, C2H4+,C2Hj+,and C3Hs+do not react appreciably with methane even though exothermic reactions can be written f o r the latter three ions. The rate constant f o r reaction of C2HZi with C H , is jive times the rate constant f o r reaction of CzH3+ with CH,. These observations of chemical effects on reaction rates are not explained b y present theory. These data support the suggestions that the high molecular weight polymers formed in high energy irradiation of methane by ionic processes come f r o m reactions of the hydrogen-deficient ions of low concentration, C+ and CH+. The gaseous ionic reactions occurring in methane have been investigated by numerous workers. The trend in these studies has been to work at higher and higher source pressures, and as the source pressure is increased the degree of complexity of the reactions occurring and the extent of knowledge about ionic reactions have increased also. As a continuation of (1) V. L. Tal’roze and A. L. Lyubimova, Dokl. Akad. Nauk S S S R , 86, 909 (1952). (2) (a) D. P. Stevenson and D. 0. Schissler, J . Chem. Phys., 23, 1353 (1955), (b) D. 0. Schissler and D. P. Stevenson, ibid., 24, 926 (1956). (3) F. H. Field, J. L. Franklin, and F. W. Lampe, J . Am. Chem. SOC., 79, 2419 (1957). (4) V. L. Tal’roze and E. L. Frankevich, Russ. J . Phys. Chem., 34, 1275 (1960). ( 5 ) R. Fuchs, Z . Naturforsch., 16a, 1026 (1961). (6) S. Wexler and N. Jesse, J . Am. Chem. Soc,, 84, 3425 (1962). (7) F. H. Field, J. L. Franklin, and M. S . B. Munson, ibid., 85, 3575 (1 963). (8) G. A. W. Derwish, A. Galli, A. Giardinl-Guidoni, and G. G . Volpi, J . Chem. Phys., 40, 5 (1964).
Field, Munson
this trend, we wish here to report our study on the ionic reactions in methane at pressures up to 2 torr in the source of the mass spectrometer. The most recently published studies report measurements at source pressures up to about 0.3 torr, and, making reasonable assumptions about the cross section and the mass spectrometer ionization chamber dimensions, one calculates that an ion formed in the electron beam will make 5-10 collisions with gas molecules in traversing the distance between the electron beam and the ion exit slit. If the pressure in the source is increased to 2 torr, the number of collisions made by an ion within the reaction chamber increases correspondingly to 30-70, and one approaches more closely than previously to macroscopic conditions. One can expect to observe high-order reactions, slow reactions, and reactions of ions with impurities or small amounts of added substances. We shall report reactions of ions from methane with added substances in subsequent papers. On the basis of the studies made at source pressures up to 0.3 torr, it is generally agreed that the following are the most important ionic reactions occurring in methane. Primary ions CHI
+ e +CHI+, CHI+, CH2+, CH+, C+, Hz+,H+
(1)
Secondary reactions CHI+ CH,+
+ CHI + C H 2 + CH, + CHI CzHs+ + Hz
+ CHd + C ~ H I ++ HP CIHI+ + Hz + H CH+ + CHa +CzHz+ + Hz + H
CHz+
I Mass Spectrometric Studies of Methane at 5 2 Torr
(2) (3) (4) (5)
(6)
3289
Tertiary reactions CH3+
+ 2CH4 + 'C2Hi' + CHa C3H7+ + 2H2
C2H3+
+ CHd +C3Hj- + H1
(7) (8) (9)
Product ions of mass higher than that of C,H,+ are found in the methane mass spectrum, but a difference of opinion6t7exists as to whether the ions observed are truly formed as a result of gaseous ionic reactions with methane. The relative concentrations of the various ions formed at several pressures are given in Table I, which will be discussed later.
Experimental The experiments were made with the Esso (formerly Humble) chemical physics mass spectrometer, which has previously been d e ~ c r i b e d . Two ~ ~ ~ modifications were made for the measurements reported here. The more important is that the openings in the ionization chamber are made smaller to decrease gas flow and thus permit the experiments to be done at higher pressures. The new dimensions of the electron entrance slit are 0.05 X 3 mm., and those of the ion exit slit are 0.05 X 5 mm. The dimensions of the analyzer entrance slit are 0.25 X I O mm., and an attempt to use a very narrow slit (0.05 X 10 mm.) to maintain a low analyzer pressure was unsuccessful. The differential pumping in the system is such that at a source pressure of 2 torr, the analyzer pressure is about 2 X torr. The other modification made in the equipment comprises an improvement in the method of measuring the pressure in the ionization chamber. A 4-mm. 0.d. glass tube has been inserted down the inside of the gas entrance tube, and the end of this tube reaches to a point just behind the ion repellers in the ionization chamber. The other end of the tube passes out of the inlet line through a ring seal and is connected to a McLeod gauge. Thus the McLeod gauge measures the pressure in the ionization chamber directly. Direct measurement of the source pressure has been achieved previously. la Checks were made between the pressures determined by the previously utilized thermistor techniqueg and the pressures determined directly with the McLeod gauge. Excellent agreement (on the order of 1-273 was found between the pressures determined by the two methods. The ionizing electron current was about 0.05 Ma. measured with no gas in the ionization chamber. The electron current emitted from the Ir filament was maintained constant during each experiment. To improve the penetration of the electrons into the ionization chamber at high pressure, the electron voltage was maintained at 150 v. in the course of these measurements. The repeller voltage was maintained at 2.5 v. (6.25 v./cm.), and the ion-accelerating voltage was maintained at 3000 v. All measurements here reported were made with the metastable suppressor maintained at a high enough potential to suppress completely the (9) F. H. Field and M. S . B. Munson, paper presented at the 11th ASTM Conference on Mass Spectrometry, San Francisco, Calif., May 1963. ( I O ) (a) V. L. Tal'roze and E. L. Frankevich, Russ. J . Phys. Chem., 34, 1275 (1960); (b) S . Wexler and R. Marshall,J. A m . Chem. Soc., 86, 751 (1964).
3290
collection of collision-induced metastable ions. The source temperature was maintained at 210 + 10". It was found that the distribution of ions to be found in methane at high pressure depends very strongly on the presence of even quite small amounts of certain impurities (especially water and ethane), and special precautions had to be taken to ensure that reproducible, meaningful results were obtained. To try to understand early unreproducible results, a systematic variation of instrumental parameters was made which gave essentially no changes in the ionic distribution. A n appropriate amount of Phillips research grade methane was condensed in a trap with liquid nitrogen, and then a center cut was distilled onto Linde 5A molecular sieve maintained at liquid nitrogen temperature. A center cut was distilled from the material sorbed on the molecular sieve and stored in an appropriate glass vessel. The apparatus, including the molecular sieve, was evacuated for 12 hr. prior to use, and immediately before the purification operations the molecular sieve was heated to 350-400" and evacuated for 15-30 min. The temperature of the glass storage vessel was also increased during evacuation to reduce sorption of impurities on the wall. The total ion current passed through a maximum at 0.2-0.5 torr and then decreased continuously as the pressure was increased. Material sorbed on the interior surfaces of the mass spectrometer and the gas inlet lines also caused difficulties, and evacuation of the mass spectrometer for several days after exposure to atmospheric pressure was imperative if reliable measurements were to be made. In fact, really satisfactory measurements were obtained only after the mass spectrometer had been in continuous use in CH, service for several weeks. The presence of water as an impurity could be deduced from the appearance of H 3 0 + (m/e 19) in the mass spectrum, and similarly C2H7+ (m/e 31) probably was indicative of the presence of ethane. At a source pressure of 0.2 torr, H 3 0 + rather than H 2 0 +should be proportional to the water concentration, and it was estimated that the water concentration was of the order of 0.01 %. The ethane concentration is not reliably known, but it is probably about the same as that for water. As a matter of interest, although the water is present in small concentration, at a source pressure of 2 torr the H,O+ ion comprised about 1.5 % of the total ionization in the system, thus providing an illustration of the high probability of proton attachment to water and the relatively large number of collisions the ions make in the ionization chamber at high pressure. Results Table I contains the relative intensities (assumed to be the same as relative ion concentrations at the ion exit slit) observed at three values of the ionization chamber pressure in a typical experiment. Also included in Table I, as a matter of interest, are the relative concentrations of primary ions in methane taken from the literature. The ions tabulated at any given pressure comprise about 99.95 % of the total ionization recorded on the chart issuing from the mass spectrometer. Among those omitted are ions with mass between 71 and approximately 125, which are of very small relative concentration, that is, on the order of
Jorm~crlof the American Chemical Society / 87:15 1 August 5, 1965
0.6I
HI
IO7
0.5
t
I
I
I
I
0.21
I
0
I
0.50
I .oo
P(CH41
Figure 1. Relative concentrations vs. ionization chamber pressure of CH4.
or less. The variations with ionization chamber pressure of the relative concentrations of most of the important ions are shown in Figure 1. The experimental points are omitted in this figure to save space and reduce confusion, but to illustrate the experimental precision, we give in Figure 2 a pressure plot of the relative concentration for C2H5+with the experimental points included. This is reasonably typical, although for a few low concentration ions the experimental scatter was greater. Table I. C H 4Mass Spectra at Various Pressures (torr)“ Rel. intensity, Zi/ZZt at
7 -
7
~
Pa “= m/e l e 4 b
12 13 14 15 16 17 19 26 27 28 29 31 39 40 41 43 55 57 69 71
0.013 0.038 0.076 0.40 0.47 0.005
P, = 0.34
P.
2.00
TORR
Figure 2. Relative concentrations of C2Hs+vs. ionization chamber pressure of CH4. The different symbols represent replicate runs made over a period of about 2 months.
Ps (CH41, TORR
Ion
I
1.50
1
=
0.91
7 . 7 x 10-6 2 . 2 x 10-4 4 . 1 x 10-6 7 . 7 x 10-4 1 . 3 x 10-4 8 . 3 x 10-4 0.0061 6 . 0 x 10-4 0.0040 0.478 0.351 0.0038 0.0022 4 . 7 x 10-4 0.0051 0.039 0.020 0.032 0.406 0.440 8 . 5 x 10-4 0.0012 5 . 8 x 10-4 0.0016 0.0016 0,0029 0.055 0.086 0.0071 0.0047 6 . 3 x 10-4 7 . 0 x 5 . 5 x 10-4 0.0013 7 x 10-6 4 x 10-6 4 x 10-6 6 X le6
P.
=
,
1.9
.
.. ...
...
0.452 0.030 5 . 3 x 10-4 0.024 0.349 0,0055 3 . 6 x 10-4 0.0011 0.053 0.028 0.0018 0 ,0051 0.0011 ...
0 Field strength (FS) = 6.25 v./cm. ; electron energy (EV) = 150 v. Taken from “Catalog of Mass Spectral Data,” A.P.I. Research Project 44, Chemical Thermodynamics Properties Center, Texas A & M University, College Station, Texas (EV = 70 v.). c Intensities given are those of the higher mass (hydrocarbon) component of the doublet observed at mass 31.
CH5+ and CzH5+. The CH4+ and CH3+ ions are the dominant ions in the primary mass spectrum of methane, comprising about 87% of the total primary ionization. These ions react with methane according to reactions 2 and 3 to form the CH5+ and C2H5+ ions as products, but, as may be seen from Figure 1 Field, Munson
and Table I, these product ions are stable in methane, for no perceptible change in the ionic concentration occurs with increasing methane pressure. If one assumes that the minimum detectable change in relative intensities of these ions is about 0.05, we deduce from the observed results that the rate constants for the reactions of CH5+ and C2H5+ with methane are smaller than approximately 10- l 2 cc./molecule sec. This nonreactivity of C2H5+ is of interest since an exothermic reaction can be written for a process consuming C*Hb+, namely GHs+
+ CH4 +sec-C3H7++ HZ
(AH = -17 kcal./molell) (10)’l.
The constancy of the relative intensities of CH5+ and C2H6+at pressures above 0.25 torr is in disagreement with the results of previous studies on methane. Wexler and Jesse6 found that above a source pressure of about 0.2 torr the relative intensity of C2H5+ increased steadily, but that of CH5+ passed through a maximum and then decreased. It was suggested that a reaction with CH, converting CH5+ to C2H5+occurs. By contrast, we have previously found? that above a pressure of about 0.15 torr the CH5+ relative intensity continuously increased whereas the C2H,+ intensity tended to decrease somewhat, and we suggested that a process forming CH5+ at the expense of C2H5+was occurring. In both of these contradictory studies the maximum ionization chamber pressure achieved was 0.35-0.5 torr, and the pressure range over which the conversion of C2H5+into CH5+ or vice versa could be observed was relatively limited. In the present work no net conversion of these two ions is observed over a pressure range of approximately 1.7 torr, and we believe that this finding is more reliable than those of either of the two previous works. In a subsequent paper we will consider the reactions of CH j+ with other compounds, which may have been impurities in the methane used by Wexler and Jesse6 and which can explain their results. We have no explanation for the error in our previous work,? except f o r some instrumental effects. (1 1) Heats of formation used in this paper are taken from the compilation given in F. H. Field and J. L. Franklin, “Electron Impact Phenomena,” Academic Press Inc., New York, N. Y . , 1957. ( l l a ) NOTEADDEDIN PROOF. As will be discussed later, this reaction may be responsible for the small amount of CIH,’ actually observed, but it may still be asserted that CsH$ is effectively unreactive in methane.
1
Mass Spectrometric Studies of Methane at
5 2 Torr
3291
Interesting comparisons can be made between the present work and the radiation chemistry studies of methane carried out by Ausloos and co-workers. 1 2 , 1 3 From isotopic analyses of the products formed in the irradiations of mixtures of CH, and/or CD, in the presence of various additives, particularly perprotonated or perdeuterated higher hydrocarbons, these workers make deductions concerning the ionic reactions occurring in the system and also the radiation yields of certain ionic intermediates. Using this technique, it is found that even with as little as 0.01 of added C3H, or G H I o , the majority of the ethyl ions formed in methane by reaction 3 are consumed by undergoing a hydride ion transfer reaction with the additive to produce ethane. If as an approximation we consider the cross section for the collisions of an ethyl ion with methane and propane to be equal, these results mean that on the average the ethyl ion survives approximately 10,000 collisions with methane molecules before reacting with the added propane or butane. Our results require that no reaction of the ethyl ion occurs in 50-100 collisions with methane, and thus both results indicate that ethyl ions react very slowly, if at all, with methane. Other evidence on this point is given by Ausloos and co-workers, and they also present evidence that the conversion of CH6+to CzH6+postulated by Wexler and Jesse' does not occur. We consider that with regard to the reactions of CHj+ and C2H6+ the results of Ausloos and co-workers and the present results are mutually corroborative. CzH4+. The CzH4+ion is formed from methane by reaction 4, and, as may be seen from Table I and Figure 1, little or no further reaction of the product ion with methane occurs. The experimental scatter for this ion was for unknown reasons always relatively high, but even so there is no question that the relative intensity of CzH4+at 2 torr is not significantly different from the value at 0.2 torr. Thus the rate constant for the reaction of this ion with methane is also on the order of 10-l2 cc./molecule sec. For this ion one can write an exothermic reaction with methane, namely
+
C2H4+ CH4
+C,Hs+ + Hz
( A H = -6 kcal./mole) ( 1 1 )
CzH3+ and CzH2+. The CzH3+ ion (m/e 27) is formed in methane by reaction 5, and, as can be seen from Figure 1 and Table I, it is consumed by further reaction with methane. Little doubt exists6,' that C3H5+is the product formed from CzH3+ according to the reaction 9. If this be the case, in the absence of loss processes one would expect the sum of the relative concentrations of C2H3+and C3H5+to be independent of the pressure of methane, but an inspection of Figure 1 shows that in fact this is not the case. We must conclude that approximately half the C2H3+ ions initially formed are lost at the higher pressures, but we are unable to account for this loss. It might be thought that a subsequent reaction of C3H5+ occurs, but no higher molecular weight product ion of concentrations comparable to the intensity loss in C&3+ and C$H5+ is found. From Figure 1 it may be seen that at higher pressures (above the concentration maximum) the concentration (12) P. Ausloos, S. G. Lias, and R. Gorden, Jr., J . Chem. Phys., 39, 3341 (1963). ( I 3) P. Ausloos and S. G. Lias, ibid , 38, 2207 (1963).
3292
Journal of the American Chemical Society
of GHa+ exhibits an exponential decay corresponding to pseudo-first-order kinetics. Thus we have made some logarithmic plots of the relative concentration, and a typical example is shown in Figure 3. The C2H2+ ion is formed in methane by reaction 6 and reacts further, presumably to give C3H3+,6s7 according to the reaction
+
C Z H ~ + CH4 --+C3H,+
+ Hz + H
(AH
=
-34 kcal./molc) (12)
although other products have also been reported. The concentrations of these ions are too small to be included conveniently in Figure 1, but we include in Figure 3 the semilogarithmic plot of the decay of the relative concentration of C2Hz+. The slopes of the linear portions of the lines are equal to k7, the products of the rate constants for the reactions and the residence times of the ions in the ionization chamber. Values obtained for the two ions are given in Table 11 (along with previously quoted rate constants), and the uncertainties are the average deviation from average of the replicate determinations. Table 11. Reaction of Product Ions in Methanc
Ion
a
sec.
kT,
CC./mOlcCUlt.
CHj' C2Hs'
< 2 X IO-'' (2 X IO-'' 1 . 2 x IO-"*
CzHa+ CyHzCZH9+ C,Hi+
< 5 X lo-'' 6.1 0.9 X 1 . 2 zk 0 . 2 X lo-''
*
Small
k , cc./molecule sw."
< 10-19 < 10-1z 0 . 9 x lo-'**
